On 17/11/2020 09:57, Sole Galli via scikit-learn wrote:
And I understand that it has to do with the cost function, because if we
re-balance the dataset with say class_weight = 'balance'. then the
probabilities seem to be calibrated as a result.
As far I know, logistic regression will have well
I am not sure if you are using "calibrated" in the correct sense.
Calibrated means that the predictions align with the real world
probabilities.
so if you have a rare class it should have low probabilities
On Tue, Nov 17, 2020 at 9:58 AM Sole Galli via scikit-learn <
scikit-learn@python.org> wro
Hello team,
I am trying to understand why does logistic regression return uncalibrated
probabilities with values tending to low probabilities for the positive (rare)
cases, when trained on an imbalanced dataset.
I've read a number of articles, all seem to agree that this is the case, many
show
